Comparison

NORBU vs. Eve

Eve is an AI demand letter product that learns a firm's voice from its own past demands. NORBU is a demand intelligence layer that examines the assembled file before the demand is sent. This page compares the two on demand readiness, medical chronology and demand package completeness, using material Eve publishes about itself.

How this page is sourced

Every Eve statement on this page paraphrases material Eve publishes on its own website, reviewed on 31 August 2026. None of it is a benchmark, a test result or a claim about behaviour Eve has not documented. Where a capability is not publicly documented, this page says so — it does not assert the capability is absent. Vendor marketing changes; verify current capabilities with Eve directly before making a purchasing decision.

Eve

Eve markets AI demand letters that "grow your demand output without growing your team" — drafts in the firm's own voice, trained on the firm's completed demand letters rather than generic templates, grounded in the case file and ready for attorney review.

NORBU Demand

NORBU Demand is demand intelligence: a pre-send examination of the complete file that scores readiness, reconciles damages, cites every assertion to a source page and reports what is missing before the package leaves the firm.

Dimension one

Demand readiness

Eve's published objective is demand output — more demands, in the firm's voice, with less attorney time. Readiness is a different question: whether this particular file should be sent at all yet.

DimensionEve — as publishedNORBU Demand
Primary orientationDrafting. Eve publishes demand generation in roughly 10–15 minutes, reduced to 3–5 minutes of human time with its drafting agent.Examination. The readiness review reports on the file's condition before anything is sent.
Post-draft review burdenEve publishes that drafts arrive roughly 85–90% ready, with review for accuracy and bracketed case-specific details expected.That final pass is the product. Unsupported assertions, unreconciled figures and gaps are reported as findings, not left to the reviewer to notice.
Readiness scoringNot publicly documented as a scored output. Eve publishes gap analysis inside its Medical Overview.A scored readiness analysis across coverage, support, reconciliation and exposure.
Weaknesses an adjuster will raiseEve publishes inline refinement — highlight a section and ask Eve to sharpen tone, add supported facts or tighten an argument.Surfaced as findings before send, for attorney judgement, rather than as editing suggestions.
Reviews a demand drafted elsewhereNot publicly documented. Eve's demand capability is described as drafting within Eve.Yes. The review operates on the package and the records behind it.
Firm voice and toneCentral to the product. Trained on the firm's own completed demands, including letterhead and headers.Not a NORBU claim. NORBU does not market voice matching.

Eve optimises the 85–90%. NORBU is built for the remaining 10–15% — the part where an under-documented specials figure or a missing set of records changes what the case is worth.

Dimension two

Medical chronology

This is the dimension where the two products come closest. Eve's Medical Overview and NORBU's chronology both turn records into a structured, dated account of treatment.

DimensionEve — as publishedNORBU Demand
Chronology outputPublished. Eve's Medical Overview produces a chronology from uploaded records and bills.A cited chronology with provider, date of service and source page on every entry.
Gap analysisPublished as part of the Medical Overview.Published. Gaps are reported with the dates and the bracketing records, and flagged against what the demand asserts.
CodingPublished. Eve's Medical Overview includes ICD codes.NORBU records what the documents state and cites it. It does not add clinical interpretation.
Relationship to the demandThe Medical Overview feeds the demand narrative — Eve publishes that running it first gives the strongest foundation.The chronology is the standard the demand is checked against. Assertions that the chronology does not support are reported.
Missing recordsNot separately documented as a distinct output; gap analysis is published within the Medical Overview.Missing records are reported explicitly as a completeness finding before send.

Eve builds the chronology to write from. NORBU builds it to check against. Both are legitimate; they answer different questions about the same records.

Dimension three

Demand package completeness

Eve publishes exhibit handling as part of the demand package. The distinction is between assembling a package and auditing one.

DimensionEve — as publishedNORBU Demand
Exhibit handlingPublished. Attach records, bills and photos; reorder, adjust page ranges, rename and export as a combined PDF or separate files.Automatic indexing, with each exhibit tied to the assertions in the letter that depend on it.
Damages reconciliationPublished. Eve's Medical Overview includes damages calculations.Line-by-line reconciliation of billed, adjusted and paid amounts against the underlying records.
Unsupported-assertion checkEve publishes that drafts are grounded in the case record and that bracketed items are flagged for the reviewer to complete.Every assertion is matched to a source page; anything unmatched is reported as a finding.
Case-type coveragePublished across auto and trucking, premises, dog bite, employment, workers' comp, med mal and wrongful death.Readiness review is case-type aware, but NORBU is built for personal injury demand packages specifically.
Commercial modelNot published on the site. Eve directs prospective firms to schedule a call.Pay-per-demand, with a free entry-level product for firms that want to run a file first.

Eve assembles a package as the end of drafting. NORBU treats the package as the thing to be examined — the completeness report is the deliverable.

In fairness

Where Eve is strong.

A comparison that only flatters the author is not a comparison.

  • Voice. Eve trains on a firm's own completed demands and publishes that it learns tone, structure, level of detail and letterhead — a genuinely hard problem, and one NORBU does not claim to solve.
  • Per-case-type agents. Eve publishes separate drafting agents for MVA, trucking, premises, dog bite, employment, workers' comp, med mal and wrongful death, and for specific carriers or jurisdictions.
  • Medical Overview. Eve publishes a Medical Overview that produces chronology, damages calculations, ICD codes and gap analysis, and feeds the demand narrative.
  • Automation. Eve publishes a Demand Drafting Agent that begins drafting when new documents arrive rather than waiting for a user to press a button.

Eve and NORBU sit at different moments. Eve's published answer to "how much editing is required" is a draft that is roughly 85–90% ready, with review for accuracy and bracketed case-specific details still expected. NORBU exists for exactly that remaining review — the pass that decides whether the package is defensible, on a demand drafted anywhere.

Questions

What firms ask.

Is NORBU a replacement for Eve?
They solve adjacent problems. Eve produces demands in a firm's voice; NORBU examines a demand package before it is sent. A firm that already likes its drafts may still want an independent readiness check on the file behind them.
Can NORBU review a demand that Eve drafted?
Yes. The readiness review runs against the assembled package and the underlying records, not against the drafting tool.
Does NORBU write demands in my firm's voice?
No. Voice matching is not something NORBU markets. NORBU's output is an evidentiary and completeness analysis of the file.
How current is this comparison?
Every Eve statement above paraphrases material published on eve.legal and was reviewed on 31 August 2026. Product marketing changes. Verify current capabilities directly with Eve before making a purchasing decision.

Sources

Where these Eve statements come from.

Eve and its product names are trademarks of their respective owner. NORBU Demand is not affiliated with, endorsed by or sponsored by Eve.

NORBU Demand provides analytical support for counsel. It does not determine legal value, predict settlement outcomes, or provide legal advice. Every observation is presented for attorney review.

Test it against your current process.

One file, one readiness review.